BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:0tvdfjtq0kihvbp2br6eru9uio@google.com
CATEGORIES:RTportal.ru
CREATED:20170606T092329
SUMMARY:Семинар СМиГО: Multi-Class Classification: How to Deal with Multi-class Huge Scale Problems Efficiently?
LOCATION:ИППИ РАН в 615 аудитории (6 этаж)
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:Presenter: Yuri Maximov (Skoltech, Los Alamos National Laboratory)\n\nTitle
 : Multi-Class Classification: How to Deal with Multi-class Huge Scale Probl
 ems Efficiently?\n\nAnnotation:\n> Multi-class classification problems are 
 motivated by a numerous applications in \n> informational retrieval, text a
 nalysis and engineering systems. Nowadays \n> problems require time and mem
 ory efficient algorithms with strong statistical \n> guarantees. In our tal
 k we focus on a line of multi-class learning strategies \n> proposed in a s
 eries of recent works joint with MR Amini, Z Harchaoui, \n> F. Iutzeler,  B
 , Joshi, I Partalas and  D Reshetova. Finally we illustrate the \n> efficie
 ncy of our algorithms on several practical problems arises in text analysis
  \n> and energy systems.  This event was imported from: https://rtportal.ru
 /index.php/kalendar-sobytij/eventdetail/124/-/seminar-smigo-multi-class-cla
 ssification-how-to-deal-with-multi-class-huge-scale-problems-efficiently?tm
 pl=component
DTSTAMP:20260812T174858Z
DTSTART;TZID=Europe/Moscow:20170301T183000
DTEND;TZID=Europe/Moscow:20170301T200000
SEQUENCE:1
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR